Skip to main content
Premium Trial:

Request an Annual Quote

Genentech Basket Trial Data Supports Genomic Profiling of Salivary Cancer Patients for Precision Treatment


This article has been updated with comments from Genentech.

NEW YORK – The latest data from Genentech's MyPathway basket trial suggests that patients with rare salivary gland cancers should be genomically profiled to give them the opportunity to receive molecularly guided targeted treatment instead of more toxic chemotherapy or radiotherapy.

"There are [cancers] where there can be more debate" about the need for genomic profiling, "and there are [cancers] where there's less debate," said Razelle Kurzrock, director for the Center for Personalized Cancer Therapy at the University of California, San Diego, and the lead author of the latest MyPathway analysis published in the Annals of Oncology. The study was done in a small number of patients, Kurzrock acknowledged, but given the rarity of this tumor type and the lack of good treatment options, she believes that the response rates seen in this analysis makes a strong case for genomically profiling patients with salivary gland cancers. 

The data from basket studies like MyPathway is helping drugmakers understand the activity of their drugs in rare, molecularly defined patient subgroups that have been challenging to study in randomized, controlled trials. These studies are also producing data that oncologists, molecular tumor boards, and guidelines bodies can use to identify treatment opportunities for patients who lack options.

Salivary gland cancers are a classic example. These types of cancers affect 16 in 1 million individuals and are typically treated with surgery and radiotherapy. There is no standard-of-care systemic therapy. Patients treated with toxic chemotherapies are often unresponsive, and approximately 40 percent of patients with advance forms of the disease live for five years.

However, salivary gland cancers often harbor molecular alterations. For example, studies show that around 30 percent of patients have HER2 altered in some way. Basket studies like MyPathway are providing patients the chance to receive targeted therapies based on these molecular features.  

In the Phase IIa MyPathway trial, Genentech is exploring the activity of its approved drugs in new molecularly defined indications. Patients with salivary gland cancers were enrolled at a dozen sites around the country and had their tumors genomically profiled according to the tests that their physicians ordered.

As of the data cutoff in January 2018, 19 patients with aggressive salivary gland cancers were enrolled in the study: pertuzumab (Perjeta) and trastuzumab (Herceptin) if they had HER2 overexpression, amplification, or mutations; vemurafenib (Zelboraf) if they harbored BRAF mutations; vismodegib (Erivedge) if they had mutations in PTCH-1 or SMO; and the immunotherapy atezolizumab (Tecentriq) if they had tumor mutational burden of more than 10 mutations per megabase.

In the study overall, 14 of the 19 patients, or 74 percent, saw a clinical benefit experiencing tumor shrinkage or stable disease lasting more than four months.

The most frequent tumor marker, seen in 15 patients, was HER2 overexpression or amplification. In this subset, nine patients, or 60 percent, saw their tumors shrink on the pertuzumab/trastuzumab combination, with one experiencing a complete response. On average, patients with HER2 overexpression who responded continued to benefit from treatment for more than nine months, and five of these patients were still responding at data cut off. Median progression-free survival was 8.6 months, and by data cut off nine patients had died, resulting in a median overall survival of 20.4 months.

The study enrolled four other salivary cancer patients: one with a HER2 mutation, another with a PTCH-1 mutation, a third with a BRAF mutation, and a fourth with high tumor mutational burden. The HER2-mutated patient (who didn't have HER2 overexpression or amplification) had stable disease on the pertuzumab plus trastuzumab combination and progression-free survival of 11 months.

Progression-free survival was 14.3 months for the PTCH1-mutated patient treated with vismodegib and 18. 5 months in the BRAF-mutated patient who received vemurafenib. These two patients saw their tumors shrink for around 11 and 15 months, respectively, but discontinued by data cut off because their tumors started growing again. At data cut off, the patient with high tumor mutational burden who was receiving atezolizumab monotherapy saw tumor shrinkage after having stable disease for 5.4 months and continued to remain on treatment.

These results are encouraging, according to Kurzrock and colleagues, because not only do they demonstrate that salivary cancer patients are benefitting from targeted therapy, but they also suggest that these patients can avoid the toxicities of chemotherapy. For example, a recent Phase II trial looked at how HER2-positive salivary duct cancer patients fared on trastuzumab and docetaxel and reported high response rates but also grade 4 decreased neutrophil count in more than half the patients. 

Chemotherapies tend to decrease this type of white blood cell count, which makes patients vulnerable to infections, Kurzrock explained. In comparison, in the MyPathway analysis, patients tolerated the molecularly guided targeted therapies and immunotherapy well, with only one grade 3 adverse event related to study drugs and no grade 4 toxicities.

Because the drugs in MyPathway are already FDA approved in other indications, their safety profile has been studied in thousands of patients. The adverse events seen in this analysis were consistent with what has been previously seen, Kurzrock said.

Several years ago, drugmakers like Genentech and Novartis began launching basket trials to gain an early view of treatment indications that may have activity in molecularly defined patient populations and those that don’t. More recently, drugmakers have employed basket study designs to explore tissue-agnostic indications and even submitted data to the US Food and Drug Administration in regulatory submissions.

For example, Loxo Oncology’s registrational study for selpercatinib is a basket trial, which tests the drug in patients with different types of cancers that share the same type of RET mutation. The FDA is expecting data from basket and umbrella trials to only increase in drug development and regulatory submissions. 

Within MyPathway, Genentech has enrolled approximately 650 patients in seven treatment arms with eight agents. The benefit with basket studies is that they can be adapted quickly, to add or retire treatment arms, based on responses seen in patients. In the course of the trial, Genentech has closed some treatment arms due to the lack of clinical benefit in certain molecular indications, but also adapted the study to include other exploratory biomarkers for evaluation, particularly in indications where patients lack standard of care options, a company spokesperson said.

In 2018, researchers reported that between April 2014 and November 2016, approximately 250 patients with 35 different tumor types were enrolled in the various MyPathway arms. In the 230 patients who were evaluated for response or discontinued for evaluation, 23 percent of patients with 14 different types of cancer saw their tumors shrink. As part of that analysis, researchers highlighted that 38 percent of colorectal cancer patients with HER2 overexpressing or amplified tumors responded to dual anti-HER2 inhibition, while 43 percent of non-small cell lung cancer patients with BRAF mutations saw their tumors shrink with vemurafenib.  

Other researchers have also explored these indications and seen similar results. For example, in one study published in JCO Precision Oncology last year, researchers from MD Anderson Cancer Center and Lexicon Pharmaceuticals evaluated the efficacy of HER2-targeted treatment in a variety of solid tumors with HER amplification, including salivary and colorectal cancers. When patients with tumor types for which HER2 targeted therapy is currently not approved received these drugs, they lived significantly longer than those who received other kinds of treatments.  

In another basket study, also led by MD Anderson Cancer Center's Vivek Subbiah and published in JCO Precision Oncology last year, researchers reported an objective response rate of 37 percent and a median overall survival of 15.4 months in around 60 NSCLC patients with BRAF V600 mutations treated with vemurafenib. 

The strongest data to emerge from this latest MyPathway analysis is the response seen in HER2-positive salivary gland tumors on the pertuzumab/trastuzumab regimen. There is also a compelling biological rationale for the benefit seen in this group that Kurzrock observed a decade earlier while she was at MD Anderson Cancer Center.

She and her colleagues conducted studies in which they started noticing similarities between salivary duct tumors and breast tumors. Both cancers start in the tubes that carry saliva and milk, respectively, and these tumor cells look similar under a pathologist's microscope. "I thought that it was interesting that not only do they look similar, but it turns out they have some similarities in what [genes] they express," Kuzrock said.

Both types of tumors overexpress HER2. And like breast cancers, salivary cancers often overexpress estrogen and androgen. Salivary duct tumors have also been observed to harbor PIK3CA mutations, which occur in breast cancer.

Kurzrock and her colleagues were curious about the similarities they observed between these cancers, especially since Genentech's trastuzumab was already well established in HER2-positive breast cancer by then. Now, "the [MyPathway data] sort of confirms that these patients respond to these [HER2 targeted] treatments," she said.

Whether the data from MyPathway and other studies will convince Genentech to pursue approval for the pertuzumab/trastuzumab combination in HER2-positive salivary gland cancers is not yet known. The Genentch spokesperson said that the data in salivary gland cancers is encouraging, but still early, and that the biotech is continuing to evaluate the activity seen with pertuzumab/trastuzumab in further studies.

"We have a lot of data here," Kurzrock said. "Personally, I would be very happy if they took it to the FDA."

Even if they don’t lead to FDA-approved therapies, the data from these basket studies add to the body of knowledge that oncologists can use to treat their patients in the era of precision oncology. Kurzrock, who runs a molecular tumor board, said that for patients with rare cancers, oncologists often make treatment decisions in the absence of any clinical data. Basket trials fill this data void by providing doctors some clinical data to go on.

"In addition to molecular tumor boards, [these data] can also inform [National Comprehensive Cancer Network] guidelines," Kurzrock said, noting the NCCN has put in guidelines treatment data that it feels is compelling even though it didn’t lead to an FDA-approved therapy. "That’s another way of getting people to pay attention to some of these results."

The Genentech spokesperson pointed out one such example. MyPathway demonstrated benefit in HER2-amplified metastatic colorectal cancer patients who received pertuzumab and trastuzumab, particularly patients with wild-type KRAS who cannot tolerate chemotherpay. After these data were published in Lancet Oncology, NCCN recommended this regimen for this subset of patients in guidelines.